Overview

The social world contains items of many kinds, including but not limited to institutions (nations, clubs), groups (races, genders), and objects (talismans, borders). This essay explores a puzzle about the essences of social items.

Some may already find it puzzling to associate social items with essences. Since antiquity, essences played important but controversial explanatory roles in philosophical and scientific theorizing. In recent times, the most controversial application of essence has been to social items. It was once common to investigate social items by investigating their essences or natures. On this approach, discovering what a nation, gender, or border is involves discovering the essence of a nation, gender, or border. The approach, however, was often entangled with problematic presuppositions about what these essences had to be. It was once presupposed, for example, that the essences of gender had to be biological. There is now widespread consensus that this presupposition, and others like it, are unscientific, philosophically unjustified, and socially unjust. This has driven many to reject social essences. Accordingly, recent discussions of essences seldom mention social items. Their focus tends to be elsewhere: on sets essentially having their members, persons or material objects essentially having their origins, water essentially being H₂O, and the like. And when social essences are discussed, it is often to deride them as harmful vestiges of unscientific and unjust ideologies.

Nevertheless, much of the resistance toward social essences is misplaced. There are compelling reasons to believe that the problematic presuppositions are dispensable. Essence can be freed from them and the problems they induce. The residual notion of essence is compatible with good science and social progress. The essences of social items may then be regarded as no less worthy of serious inquiry than the essences of sets and the like. If there is a real surprise here, then it is the prospect of learning from essences' application to a domain so often thought to be inhospitable to it.

But there is another threat to this prospect. It originates in the general question of where essence resides. Two approaches have endured since antiquity. The first, the Platonic approach, regards essences as somehow detached from the world. It is as if essences are given prior to their worldly manifestations, as with Plato's Forms. The second, the Aristotelian approach, regards essences as somehow embedded in the world. Essences are as much a part of the world as their worldly manifestations, as with Aristotle's universals.

The two approaches conflict. Their conflict, in its most abstract form, is generated by three claims: an essentialist claim that essences of a certain sort exist; an embedded claim that these essences are embedded in the world; and a detached claim that these essences are detached from the world. The embedded and detached claims are meant to capture the Aristotelian and Platonic approaches. They would not conflict if there were no essences of the relevant sort. But the essentialist claim entails that there are.

The conflict would not raise any puzzle were there not competing inclinations to regard some essences as both embedded and detached. But the competing inclinations are present, at least in some cases. They are perhaps strongest in the case of social items. In the Aristotelian spirit, not only do we construct social items, we also may seem to construct their natures. Clubs, genders, and borders did not predate us. We made them. And, it may seem, we made them what they are. It is, for example, essential to a border that it demarcates a nation's land. It is so because, somehow, we made it so. The essential fact about borders seems embedded in worldly facts about our institutions, needs, interests, and practices. But in the Platonic spirit, the essences of things do not seem up to us. Essences are like definitions or axioms. Even if we might adopt different definitions or axioms for different purposes, it is not as if we can change what the definitions or axioms are. Nor can we determine what the very natures of things are. Essences seem detached from any worldly grounds. Now we seem to have a puzzle. It is that we have conflicting inclinations to regard social essences as at once embedded in but detached from the world.

I have four aims in this essay. One aim is to clarify the puzzle. This involves refining the key notions (essence, detached, embedded) used to state it. These notions are of general interest apart from their application to social items. But I focus on that application here.

My second aim is to explore the puzzle's implications. The puzzle, once refined, is surprisingly resilient. Engaging with it promises to be of interest for what it teaches us about social essences in particular. But it also promises to be of broader interest. This is because it is a vivid case study of the general conflict between the Aristotelian and Platonic approaches to essences.

My third aim is to illustrate the fruitfulness of an unfamiliar distinction between facts, or truths. On the one hand, there are worldly facts that hold because of the circumstances. On the other hand, there are unworldly facts that hold regardless of the circumstances. This distinction concerns whether or not a fact is based in worldly circumstances, and so I call it the basal distinction. Elsewhere I have discussed the basal distinction at length and argued that it is of general interest apart from any of its potential applications (Raven Reference Raven2021a). But it gains further interest for its potential applications, such as to the philosophy of logic (Raven Reference Raven2020a). The basal distinction, as I show, helps refine the puzzle. It does so by refining the notions of detached and embedded. This illustrates how the distinction applies to essence and social metaphysics.

My final aim with this essay is to show how the distinction provides for a novel potential solution to the puzzle. The solution uses the basal distinction. Properly developing that distinction involves sorting through intricacies beyond the scope of this essay. That means my presentation of the solution here does not go beyond an outline. In particular, I do not argue for the solution or defend it against its main objections. My more modest aim is to add it to the list of options. Still, the solution should be of interest not only as a solution to the puzzle, but also as a blueprint for reconciling conflicts between the Platonic and Aristotelian approaches in other contexts.

The essay has three parts. In the first I formulate the puzzle and the three claims generating it (section 1). In the second part I explore these three claims in more detail and offer support for each (sections 2–4). In the third and final part I sketch a solution to the puzzle that makes crucial appeal to the basal distinction (section 5).

1. The Puzzle

The general form of the puzzle consists in three individually plausible but jointly inconsistent claims: an essentialist claim, an embedded claim, and a detached claim. My aim in this section is to give a clear formulation of the puzzle for the case of social items. Here my immediate concern is just to clarify these claims. I later consider how they might be motivated and defended.

The first claim is the essentialist claim. Some facts about social items seem inessential or accidental. To illustrate, suppose that Congress, women (the generic kind), and the Korean DMZ are social items. Then examples of inessential facts about them might include the following:

1. I1 Congress has 535 voting members.

2. I2 Women may vote.

3. I3 The Korean DMZ is a wildlife haven.

Other facts about social items seem essential or nonaccidental. Uncontroversial examples are not always easy to find. But we may suppose, if only for illustration, that examples might include the following:

1. E1 Congress is bicameral.

2. E2 Women are oppressed as women.

3. E3 The Korean DMZ is unfortified.

The essentiality of these facts is stated by essentialist facts. These are:

1. S1 Congress is essentially bicameral.

2. S2 Women are essentially oppressed as women.

3. S3 The Korean DMZ is essentially unfortified.

Atomic essentialist facts have the form: x is essentially F.

It is important to distinguish between facts having the status of being essential (E1–3) and facts reporting this status (S1–3). One reason to distinguish these is to allow for explanatory relations between them. For example, one might wish to say that Congress is bicameral (E1) because it is essentially bicameral (S1). If this explanatory claim is to have its intended significance, then S1 and E1 must be distinct. I return to this distinction later.

My focus is on essentialist facts about social items, such as S1–S3. I call these social-essentialist facts. And so I express the claim that essentialist facts about social items exist as:

This is the essentialist claim used in generating the puzzle. The claim needs qualifications, which I make when it is defended later.

The second and third claims are the detached and embedded claims. These are best characterized together. This is because they take opposing views of a contrast between essences being detached from, or embedded in, the world.

This contrast, or something near enough, has a venerable history. Perhaps its most traditional characterization was in terms of transcendence and immanence. On the Platonic approach, essences were thought to somehow transcend their worldly instances. By contrast, the Aristotelian approach took essences to be somehow immanent in their worldly instances.

My inclination, however, is not to characterize our contrast in the traditional way. One reason for this is that it adds excess baggage. Transcendence and immanence, as they are often understood, apply to items of some sort. If essences are universals, then it is items of this sort—universals—that are transcendent or immanent. Or if essences are particulars, such as haecceities, then it is items of that sort—haecceities—that are transcendent or immanent. Either way involves reifying essences one way or another. But the reification is unnecessary and undesirable. It is unnecessary because essentialist claims need not be about essences. For example, in accepting that Socrates is essentially human, we might only accept something about the mode of his being human. We need not thereby reify any essence. Of course, we may go on to do so, if we wish. But the point is that our essentialist claims do not require this. That makes the reification undesirable for being unnecessary.

We want to characterize the contrast between detached and embedded without reifying essences. One way is to give the characterization directly in terms of the essentialist facts themselves. This might seem to replace one reification—of essences—with another—of facts. And it might be doubted whether this replacement is beneficial. If indeed there is such a replacement, it might be argued that the reification of facts is less problematic than the reification of essences. One might argue, for example, that anyone is or ought to recognize the existence of facts whatever their views on essence might be. But it might also be argued that the appearance of a replacement is not genuine. My talk of facts is, ultimately, a useful convenience that can be translated with minimal distortion into fact-free talk. We may, for example, translate explicit fact-talk, such as ‘the fact that Congress is essentially bicameral’, into fact-free talk about truths, such as ‘the truth that Congress is essentially bicameral’. Or we may translate it into fact-free and truth-free talk, such as, ‘Congress is essentially bicameral’. Translations like these, while inconvenient, are available. Their availability ensures that indulging in fact-talk should not create new problems later.

Our approach characterizes facts as detached or embedded. We may say that whether a fact is detached or embedded is determined by its basal status. To a first approximation, the basal status of a fact concerns the basis for its obtaining. Does the fact hold because of the circumstances? If so, it has a basis in the world. It is a worldly (or circumstantial) fact, and so is embedded. Or does the fact hold regardless of the circumstances? If so, it does not have a worldly basis. It is an unworldly (or acircumstantial) fact, and so it is detached.

I illustrate the intended contrast by example. Consider the following claim about Taylor:

This is a worldly fact. A telltale sign is how T's obtaining is based on worldly circumstances, such as the time. We may sensibly ask whether Taylor was singing or now sings or will sing. And the answer will depend, accordingly, on the past or present or future circumstances. By contrast, consider the arithmetical claim:

This is an unworldly fact. A telltale sign is how Z's obtaining is not based on worldly circumstances, such as the time. It does not seem that we may sensibly ask whether zero was even or is now even or will be even. Or, at least, if there is any sense to these questions, it is only by contorting the original sense almost beyond recognition.

It is tempting to try to assimilate the basal distinction to more familiar distinctions. One might, for instance, attempt to assimilate it to the semantic distinction between analyticity and syntheticity, or the epistemic distinction between apriority and aposteriority, or the modal distinction between necessity and contingency, or various other distinctions. Elsewhere I have argued at length against such attempts at assimilation (Raven Reference Raven2020a, Reference Raven2021a). Their failure, however, does not require denying potential implications between the basal distinction and the other distinctions. But it does undermine the expectation for an analysis of it in terms of them. I therefore propose exploring the basal distinction's applications without first demanding an analysis of it. This does not rule out the possibility of an analysis later. For now, I clarify the distinction by its uses.

My first use of the basal distinction will be to help characterize my theses of embedded and detached. They are as follows:

Now the joint inconsistency of the three claims is evident. Social-Essentialism implies that there are facts about social essences. Embedded implies that they are worldly, whereas Detached implies that they are unworldly. But no fact can be both. So we have a contradiction.

Of course, that some claims are inconsistent need not itself be puzzling. What makes their inconsistency puzzling is, as I show below, that each claim is individually plausible. That is the puzzle.

This puzzle has analogues in other domains. For example, analogous considerations support essentialist, embedded, and detached claims for a puzzle about artifacts. Indeed, the analogies may extend even further to what I call constructed items. These are items constructed or produced by us, including but not limited to social items and artifacts. It seems it is possible to adapt the considerations above to generate a puzzle for constructed items in general.

Recognizing the potential extensions of the puzzle illustrates its broad interest. But just how forceful the puzzle is will be domain-specific: it will wax and wane with how plausible the generating claims are in a given domain. I focus on the social domain because it is among the most forceful. In the next few sections, I will argue that each claim generating the puzzle is individually plausible. That is not to say that they are all true. But given that the claims jointly conflict, their individual plausibility will be enough to establish that the puzzle is indeed puzzling.

2. Social-Essentialism

Doubts about Social-Essentialism arise before the puzzle does, which may only reinforce the puzzle. But before engaging these doubts, it is worth noting just how concessive we may be. For we may explore what the essences of social items would be were they to have essences, even while bracketing whether they really do. This exploration is not idle. It should interest friends and foes of essences alike because it promises to clarify what friends accept and foes reject.

We need not, however, remain so concessive. For many of the common doubts about social essences are not compelling. Some of them derive from confusions about what Social-Essentialism entails. These may be dispelled with a few qualifications.

The first qualification is that Social-Essentialism does not specify just which social-essentialist facts there are. Perhaps they include some or all of S1–S3. Or perhaps not. Of course, it would be nice to know just which social-essentialist facts there are. But Social-Essentialism itself only says there are some without specifying which. This dispels any doubts about Social-Essentialism making the wrong predictions about which social-essentialist facts there are.

The second qualification is that Social-Essentialism is silent on the source of the social-essentialist facts. Perhaps they are conferred on items by us. Or perhaps we somehow construct them. Or perhaps they were generated by natural processes. Or perhaps they obtain in a Platonic realm. Or perhaps some have one source while others have another. Or perhaps they have no source. Of course, it would be nice to know just what, if any, the sources of the social-essentialist facts are. But Social-Essentialism itself is neutral on this. This dispels any doubts about Social-Essentialism making the wrong predictions about the sources of social-essentialist facts.

Other skeptical doubts about Social-Essentialism are not so easily dispelled. There are three main kinds. The first doubts the existence of social items. The second doubts essence in general. And the third doubts the application of essence to social items in particular.

First, one might reject social essences by rejecting social items. This is not merely to say that they do not fundamentally exist or exist in reality. It is the stronger view that they do not exist at all. There are no nations, clubs, races, genders, talismans, or borders. This may be combined with the assumption that there are essential facts about social items only if social items exist. If so, then nihilism implies that Social-Essentialism is false. I will, however, just assume that social items exist. This is justified by the sheer obviousness of their existence, however controversial it may be just which there are, what they are, how we know of them, and so on. Of course, the nihilist will not be impressed by the charge that they deny the obvious. But the rest of us will see just how desperate and incredible it would be to avoid the puzzle by rejecting Social-Essentialism in the nihilist's way.

Second, one might reject social essences in particular because one rejects essences in general. Although there is a venerable tradition locating essence at the core of metaphysics, there is a competing tradition that seeks to sanitize essence, or to eliminate it altogether (Quine Reference Quine1976; Sullivan Reference Sullivan and Barnes2017). In this vein, one might say that social items in particular lack essences because all things in general lack essences. If social items have no essences, then there are no essential facts about them. And if so, Social-Essentialism is false. But the question of whether it is legitimate to appeal to essence in general reaches far beyond the immediate concern with the puzzle. Earlier I suggested that the puzzle brings into sharp relief the conflict between Platonic and Aristotelian approaches to essence. One could take the puzzle to bear on essence in general if one took these approaches to be both obligatory while unavoidably in conflict. But, so far, there is no reason to expect this outcome. And so it seems one may, at least provisionally, insulate the puzzle from its potential impact on essence in general. In this spirit, I will assume that the appeal to essence in general is legitimate.

Third, one might reject social essences in particular because it may seem as if essence cannot be properly applied to social items. In this vein, it is sometimes argued that metaphysical frameworks using notions like structure, ground, fundamentality, and essence are inhospitable to social metaphysics and feminist metaphysics (Barnes Reference Barnes2014; Mikkola Reference Mikkola2015; with replies from Schaffer Reference Schaffer2017; Sider Reference Sider2017). It must be conceded that some of these frameworks—or at least the rhetoric used to motivate them—may indeed be inhospitable. But this is not to concede that all frameworks are inhospitable. And some argue that other frameworks are, indeed, hospitable after all (Raven Reference Raven2017: 643–49; Passinsky Reference Passinsky2019, Reference Passinsky2021; Mason Reference Mason2021). At the very least, the application of essence to social items is on the cards.

Even if the application is on the cards in principle, it may still seem illicit in practice. To illustrate, consider the especially challenging case of gender. It was once common to assume that genders were essentially biological. But this is now widely rejected for being unjustified and unjust. While controversies remain over whether science supports sex being biological (Dupré Reference Dupré1993; Nanay Reference Nanay2010; Wilson, Barker, and Brigandt Reference Wilson, Barker and Brigandt2007), there is a broad consensus that it does not support gender being biological. Nor is there any plausible a priori justification for it, either. Moreover, social theorists have unmasked how the assumption can lead to various injustices (de Beauvoir Reference de Beauvoir and Pashley1984; Butler Reference Butler1988; Cornell Reference Cornell1993; Lorde Reference Lorde1984; Schor and Weed Reference Schor and Weed1994; Spelman Reference Spelman1988).

What lesson should we infer from this case? Many would infer that the application of essence to gender is illicit. But the inference is fallacious. It just does not follow that genders have no essence from their having no biological essence (Witt Reference Witt1995; Mason Reference Mason2016). A better lesson is that genders do not have biological essences, which is compatible with their having essences of another sort. In general, cases like this one rely on transferring doubts about bad views of social essences onto the social essences themselves. But the transfer is illicit and should be resisted. As a result, these cases do not support doubts about the application of essence to social items.

These considerations together suggest that doubts about social essences are weaker than they may first seem. The problematic presuppositions once foisted upon social essences are, in fact, separable from them. Freeing social essences from them reveals their compatibility with good science and social progress.

This result, however, would be of only meager interest were the appeal to social essences dispensable. But, while recognizing the potential risks of appealing to social essences, it nevertheless remains difficult—if not impossible—to do without them entirely. There are at least three reasons why.

One reason is that there is simply a strong presumption for social essences. This is clear from our ability to comprehend and our willingness to accept essentialist claims about many social items. We have, it seems, no difficulty comprehending and distinguishing between inessential claims about social items (I1–I3), essential claims about social items (E1–E3), and social-essentialist claims (S1–S3). Admittedly, not all examples of social-essentialist claims will be uncontroversial. But that is no obstacle to grasping what social-essentialist claims are supposed to be like.

A second reason is that social essences would appear to be explanatorily and ideologically fruitful (Haslanger Reference Haslanger2012; Witt Reference Witt2011; Passinsky Reference Passinsky2021). Some of their fruitfulness derives from the fruitful roles essences play in general. One of these roles is fixing the domain (Raven Reference Raven2021b). When speaking of a chemical kind (gold), a species (tigers), or a person (Saul Kripke), we might ask what it is we speak of. And we may answer by stating what the essence of the chemical kind, species, or person is. This also holds for the social. To specify a social item's essence is to fix the subject matter.

This domain-fixing role feeds into a second role for essence: specifying an item's immutable core. Once we discover the essence of a chemical kind, a species, or a person, we thereby learn that it must be so if it is to be what it is. It might change some of its inessential features and still be what it is. But were its essential features to change, it would cease to be what it is. This also holds for the social. To specify a social item's essence is to specify its immutable core.

A third role is vindicating certain projects. This involves the application of the previous roles to various subject matters. These subjects may be topical or even urgent. So there is a sense in which the third role may be of more immediate practical interest than the first two. But this third role tends to be overlooked or ignored. In any case, we may illustrate it by considering two examples.

First, consider the question of what a nation (or state) is. It is often assumed that it is essential to a nation that it have habitable territory (Crawford Reference Crawford2006). But the assumption is controversial. If, for example, climate change submerges the Republic of Maldives, will recognition of its past landedness sustain its continued existence? And landless indigenous nations may already provide counterexamples, such as with the formerly landless Little Shell Chippewa Tribe (McLaughlin Reference McLaughlin2019). These issues are important in their own right, but also for their impact on urgent legal and moral questions concerning the rights of stateless refugees and landless peoples (Alexander and Simon Reference Alexander and Simon2014a, Reference Alexander and Simon2014b, Reference Alexander and Simon2017). And answering them seems to involve determining what the essence of a nation is.

Second, consider the project of promoting social justice. One aim of this project is to clarify what gender oppression is, both to understand it and to reduce it. This plainly appeals to essence. The aim is to understand what it is to be a man or woman or genderqueer. Perhaps the apparent reference to essence can be removed or shown to be illusory. But it is unclear how this might be done without distorting the project. To illustrate, notice that Sally Haslanger (Reference Haslanger2014: 31) describes one of her projects as ‘offering a theory of what gender is’. Katharine Jenkins (Reference Jenkins2016) objects to Haslanger's (Reference Haslanger2012) theory on the grounds that it wrongly predicted that a trans woman who tends not to pass as a cis woman would not be a woman. A plausible construal of their debate is as being over the essence of womanhood (Barnes Reference Barnes2014; Passinsky Reference Passinsky2019). And so alternative construals risk distorting it. Moreover, alternative construals risk undermining the project's goal of social progress. To illustrate, consider the feminist project of opposing systematic oppression against women. Making sense of this project requires making sense of the target of the oppression they oppose. But what is it that is the target of this oppression? To answer this is to say what it is to be a woman. And so the feminist project seems to presuppose an appeal to essence. Without it, it is obscure how feminists could intelligibly formulate the oppression they oppose (see Heyes Reference Heyes1997).

There is no question that social essences raise many challenges. I do not presume to have shown that these challenges can all be met. But I have argued that it is no less challenging to dismiss social essences outright. And so it seems we should, at least provisionally, accept the appearance that there are social essences.

3. Embedded

Perhaps the main direct support for Embedded builds on the familiar idea that social items are socially constructed. It is not just that societies construct money, nations, borders, races, and genders. It is also that societies construct what they are. The social essences themselves are socially constructed. This idea is not often explicitly discussed in the literature. Still, one may find suggestive precedents. One relies on conceptual stipulations. In this vein, Amie Thomasson (Reference Thomasson2003b: 588) writes, ‘the principles [regarding sufficient conditions for the existence of institutional kinds] accepted play a stipulative role in constituting the nature of the kind’. Another precedent says essences are conferred by conceptual or explanatory practices. In the former vein, Ásta (Reference Ásta2008: 138) writes, ‘My contention here is that the property of being an essential property of an object—essentiality—is conferred. I maintain that essentiality is conferred by our use of concepts—not by how we as a matter of fact do or have used them, but how we are committed to using them’. In the latter vein, Meghan Sullivan (Reference Sullivan and Barnes2017) writes, ‘A property is only essential to an object relative to an explanatory framework, and true essence ascriptions may vary across frameworks. . . . Outside of an explanatory framework, an object's properties can be considered neither essential nor accidental’. And there are further suggestive precedents as well (Almog Reference Almog2010; Epstein Reference Epstein2015; Passinsky Reference Passinsky2020; Ritchie Reference Ritchie2015; Searle Reference Searle1995; Smith Reference Smith2001; Smith and Varzi Reference Smith and Varzi2000; Thomasson Reference Thomasson2003a).

Some might classify views like these as anti-essentialist. Indeed, some of their proponents describe their own views as anti-essentialist. But it is not necessary to make heavy weather of labels here. All the preceding views agree that there are essences but take them, in one way or another, to be constituted by our activities or practices. So they all agree that there are social-essentialist facts and that they hold in virtue of various worldly circumstances. That is, they agree that social-essentialist facts are worldly. And that just is Embedded.

There is also indirect support for Embedded. Earlier I distinguished between a fact's status and the report of its having that status. Although the distinction is somewhat unfamiliar, it is also of general interest and deserves further clarification. In addition, it has a specific application to the present context. It can be leveraged into a potential solution to the puzzle that rejects Embedded. But I will argue that the solution ultimately fails. Its failure is instructive. It indirectly bolsters Embedded and the puzzle's resilience.

The status/report distinction relies on a certain notion of a fact's status. This concerns the manner in which the fact obtains. Any manner would illustrate the idea. But modality and essence are most pertinent. To illustrate, consider Kit Fine's (Reference Fine1994: 846) famous example of Socrates and the singleton set {Socrates} containing him:

Now, S is an essential truth about {Socrates}. Following Fine (Reference Fine1994), although a necessary truth need not be essential, all essential truths are necessary. So S is also a necessary truth. But one may distinguish S from reports of its necessity and essentiality:

Because S is necessarily and essentially true, it is a necessity and an essentiality. But S reports neither of these statuses. So it is neither a modal nor an essentialist fact. By contrast, □S is a modal fact because it reports S's necessity. And ES is an essentialist fact because it reports S's essentiality. In general, we may distinguish facts having the status of being necessary or essential (necessities or essentialities, like S) from facts reporting this status (necessary facts like □S, essentialist facts like ES).

The status/report distinction may seem unstable. Isn't what it is for S to be necessary just for □S to be true? And is it not what it is for S to be essential just for ES to be true? So will the distinction not collapse?

There are, however, powerful reasons supporting the distinction's stability. For one, the distinction emerges in what is said at the level of ordinary meaning. None of S, □S, or ES could have failed to be true. But they differ in what they say. Whereas □S reports S's necessity and ES reports S's essentiality, S itself reports neither. The status/report distinction captures these contrasts in what is said.

The distinction also emerges in disagreements about what explains what. In the modal case, we may consider explanatory connections between necessities and their prejacents. On the one hand, an amodalist aims to reduce modality to regularities across possible worlds. Just as instances of a generalization help explain it, so, too, S will help explain □S. On the other hand, a modalist is content with unreduced modality. They may say that what explains S's truth is that it must be true—that is, □S. The amodalist and modalist disagree over whether S helps explain □S or vice versa. Their disagreement assumes the status/report distinction.

Analogous considerations also apply to the case of essence. To illustrate, consider an essentialist who explains essentialities in terms of essentialist facts (Glazier Reference Glazier2017). For example, given that ES obtains, then ES helps explain S. Given that nothing helps explain itself, this requires distinguishing between S/ES. And that just is the status/report distinction.

Although the status/report distinction is of general interest, it has a specific application here. It can be leveraged into a solution to the puzzle. The original formulation of the puzzle relies on:

This is a report claim. It concerns facts, like S1-S3, reporting various essentialities. Embedded says they are worldly. Given the status/report distinction, there is also a status claim about the corresponding social-essentialities:

This concerns facts, like E1–E3, that are essential as to status. Embedded* says they are worldly. The status and report claims differ. Status facts like E1–E3 concern such worldly matters as Congress and bicamerality, women and oppression, the Koreas and fortifications. So Embedded* may seem true. But report facts like S1–S3 concern essence, which is paradigmatically unworldly. So Embedded may seem false. The difference between status and report claims, however, is subtle enough that they may be confused. In particular, it may be alleged that we confused Embedded and Embedded* and thereby illicitly projected the latter's plausibility onto the former. But once the confusion is dispelled, we may accept Embedded* and reject Embedded. And because Embedded* is consistent with Social-Essentialism and Detached, the puzzle disappears.

But problems arise once one asks how social-essentialities and social-essentialist facts are explanatorily connected. Do social-essentialities help explain social-essentialist facts, social-essentialist facts help explain social-essentialities, or neither?

The last option is implausible. When a social-essentiality (like S1–S3) and its matching social-essentialist fact (like E1–E3) both obtain, then one must help explain the other. They are just too intimately related for neither to help explain the other. So let us focus on the first two options.

On the first option, social-essentialities help explain social-essentialist facts. Because we are granting Embedded*, it follows that worldly social-essentialities help explain social-essentialist facts. It is plausible, however, that if a worldly fact helps explain another fact, then that fact must also be worldly (Raven Reference Raven2021a). This implies that the social-essentialist facts must be worldly if worldly social-essentialities help explain them. And that implies Embedded.

On the second option, social-essentialist facts help explain social-essentialities. Because we are granting Embedded*, it follows that social-essentialist facts help explain worldly social-essentialities. Their worldliness may seem to imply the worldliness of the social-essentialist facts that help explain them. To illustrate, if we ask why social-essentialities like E1–E3 are worldly, we may cite their concern with worldly matters such as Congress and bicamerality, women and oppression, the Koreas and fortifications. While social-essentialist facts like S1-S3 concern essence, they also concern Congress and bicamerality, women and oppression, the Koreas and fortifications. This suggests that social-essentialist facts must be worldly if the social-essentialities they help explain are worldly. And that implies Embedded.

These considerations suggest that the proposed solution may conceal a covert commitment to the very claim, Embedded, it aims to reject. Still, the proposed solution seems appealing (especially when combined with the second option). Indeed, the solution I sketch in section 5 may be regarded as a more precise development of it. But the development relies on drawing distinctions yet to be drawn. So it would be premature to declare Embedded false and the puzzle solved. For now, the initial support for Embedded still stands.

4. Detached

There is a strong initial pull toward regarding essentialist facts as unworldly. The pull derives its strength from general considerations about essences' roles in inquiry. One role is to fix the domain. We may clarify this by contrasting it with an alternative conception of essence on which it does not fix the domain but sorts it. Shamik Dasgupta (Reference Dasgupta2016: 388) characterizes the domain-sorting conception like this: ‘one starts with the idea that there are two ways to have a property—an essential way and an accidental way—and one then takes the essentialist facts about something to be facts concerning which properties it has in the essential way. On this picture the essentialist facts are facts concerning which properties are had in that way by a given domain of things’.

The domain is given. So it is not fixed by essence. Instead, essence sorts the domain. This contrasts with the domain-fixing conception also characterized by Dasgupta (Reference Dasgupta2016: 389): ‘the essentialist facts concern what those things are in the first place. It is not that there is some independently given domain and the essentialist facts are certain facts about what properties they have. It is rather that the essentialist facts specify what the domain is in the first place’. Essentialist facts fix the domain. That is perhaps essences' primary role. It is perhaps only secondary that essentialities or essentialist facts express properties essential to items in the domain.

The domain-fixing conception comports well with the unworldliness of essentialist facts. To see why, consider how Fine (Reference Fine2005: 349) characterizes the domain-fixing conception: ‘The objects enter the world with their identity predetermined, as it were; and there is nothing in how things are that can have any bearing on what they are’. The link to unworldliness is even more explicit when Fine (Reference Fine2005: 348–49) writes, ‘the identity of an object is independent of how things turn out . . . it is the core essential features of the object that will be independent of how things turn out and they will be independent in the sense of holding regardless of the circumstances, not whatever the circumstances’.

Essentialist facts determine the identities of objects. That is how they fix the domain. But then it seems they cannot hold because of anything in the domain. And that may seem to prevent them from holding in virtue of any worldly circumstances at all. Instead, it may seem as if the essentialist facts would have to somehow obtain without regard to the worldly circumstances. Perhaps there is a way to accommodate the domain-fixing conception within the Aristotelian tradition. But it is not altogether clear how this might be done (Raven Reference Raven2021b). So the unworldliness of essentialist facts might seem to be an inevitable consequence of the domain-fixing conception.

Detaching essences from the world makes them like Plato's Forms. To the extent they are alike, criticisms of the latter may transfer to the former. For example, in Parmenides, it is worried that Forms proliferate to include such ‘undignified’ items as hair, mud, and dirt. The notorious difficulties interpreting this worry obscure how probative it is. Even so, the worry is suggestive of more pertinent worries.

One worry focuses on the proliferation of essences. If essences are detached from the world, then it seems nothing in the world could limit which essences there are. The absence of any limit suggests a proliferation of essences. It is as if all the essences there could be are already out there in a Platonic realm waiting for their chance at instantiation. While some may accept this proliferation (Yablo Reference Yablo1987; Leslie Reference Leslie2011; Inman Reference Inman2014), others find it objectionable.

Another worry focuses on the detachedness of essences. Some essences may indeed seem somehow ‘prior’ to the worldly circumstances. But others do not. In particular, social essences may seem to depend upon the worldly circumstances precisely because they are social. It may then seem objectionable, or perhaps even incoherent, to regard social essences as detached.

The objection's force can be exaggerated by neglecting the status/report distinction. Recall that the essentialist facts E1–E3 have the status of being essential while the essentialities S1–S3 report this status. Confusing them may tempt one to reject Detached for bad reasons. Consider whether Congress is bicameral, or women are oppressed as women, or whether the Korean DMZ is unfortified. These may seem to be purely worldly matters. One might conclude that Detached is false, if one confusedly took it to concern essentialities like E1–E3. But it does not. Instead, Detached concerns essentialist facts like S1–S3. And it may not seem to be a purely worldly matter whether Congress is essentially bicameral, or women are essentially oppressed as women, or the Korean DMZ is essentially unfortified. Respecting the status/report distinction allows E1–E3 to be worldly without thereby implying that S1–S3 are also worldly.

These last considerations are somewhat in tension with the considerations that concluded the previous section. There it was suggested that social-essentialist facts like S1–S3 must be worldly if the social-essentialities like E1–E3 that they help explain are worldly. The tension, however, is a manifestation of the puzzle. We have competing inclinations to regard essences as at once embedded and detached. I have argued that the tension between these inclinations will not be alleviated by blithely dismissing either of them. Instead, a satisfying solution must respect the initial support for the claims generating the puzzle even if that support is ultimately revealed as illusory. For now, the initial support for Detached still stands.

5. Toward a Solution

I have argued for each of the claims generating the puzzle. My aim was not to establish their truth but to support their plausibility. That shows that the puzzle is, indeed, puzzling. I wish to conclude by outlining a potential solution. The solution alleges that the puzzle relies on an illicit equivocation. What's novel about this solution is the specific distinction that it takes to be equivocated.

The distinction concerns how the basal status of a fact is determined. We may illustrate this by a familiar example. Recall:

On the one hand, the determination of ES may be regarded as a proximal matter. It concerns only whether Socrates is essentially a member of {Socrates} and has nothing to do with the ultimate source of this set-theoretic fact. It is, if I may put it this way, already written into the identity of the set that it has this member, and nothing in the worldly circumstances is relevant to whether this is so. On the other hand, the determination of ES may be regarded as a distal matter. It is not merely a matter of whether Socrates is essentially a member of his singleton, but also a matter of what explains, or grounds, this fact. For all that has been said, the explanation may turn out to be either worldly (for example, the fact is constructed by mathematicians’ mental acts or practices) or unworldly (for example, the fact obtains in virtue of a Platonic realm).

The example illustrates how the determination of a fact may be proximal or distal. When a fact obtains, it will or will not concern the circumstances. This is a matter of the fact's proximal basal status. But this need not engage with what, if anything, ultimately explains, or grounds, the fact (Raven Reference Raven2015; the contributions in Raven Reference Raven2020b). Its grounds, if any, will or will not concern the circumstances. And that is a matter of the fact's distal basal status. In general, a fact's proximal basal status does not involve the basal statuses of any other facts, whereas its distal basal status does.

The proximal/distal distinction enables a potential solution to the puzzle. The solution relies on the possible divergence of a fact's proximal and distal basal statuses. But the solution's viability requires refining the basal distinction. I have done so elsewhere (Raven Reference Raven2021a). Because my purposes here do not require repeating the details, below I tailor the key points to the present context.

What makes the proximal basal status of a fact proximal is its intrinsic determination by that fact itself regardless of any others. The idea is vividly illustrated by atomic facts. Let an atomic fact A be the application of an n-place property P to items x₁,…,x_n. We may then ask whether the instantiation of P by x₁,…,x_n is worldly or unworldly. Our answer determines atomic fact A's proximal basal status:

To illustrate, if we take I1 to be an atomic fact applying the property of having 535 voting members to Congress, then it is proximally worldly because it is a worldly matter that Congress has 535 voting members. And if we take E1 to be an atomic fact applying the property of bicamerality to Congress, then it is proximally unworldly because it is an unworldly matter that Congress is bicameral. In this way, the proximal (un)worldliness of an atomic fact is determined proximally by the (un)worldliness of the instantiation of its constituent property.

The question then arises of how to extend our notion of proximal basal status to complex facts. Answering this requires addressing what sorts of complexity, or structure, a complex fact may have. Will it include truth-functional complexity, allowing for conjunctive, disjunctive, or negative facts? Will it include quantificational complexity, allowing for existentially or universally general facts? Will there be other sorts of complexity too? The issues are rather delicate. A comprehensive treatment of them would take me too far afield. But it is possible to still give a penultimate gloss of the extension while reserving the right to iron out the wrinkles later.

The guiding idea is that worldliness is ‘dominant’ while unworldliness is ‘recessive’ (Fine Reference Fine2021; Raven Reference Raven2021a). This means that all it takes for a complex fact to be worldly is for it to have a worldly constituent. By contrast, a complex fact is unworldly only if all its constituents are unworldly. The guiding idea does not depend on the exact details of what a complex fact's constituents are. So it can guide the extension of proximal basal status to complex facts in abstraction from what their constituents are taken to be:

We therefore arrive at a compositional characterization of proximal basal status by combining Proximal-Atomic and Proximal-Complex. Although the characterization should not be taken as the last word on the subject, it should suffice for our purposes.

Next, consider notions of distal basal status. What makes them distal is their extrinsic determination by the basal statuses of other facts. There are many ways these other facts may exert their influence. And so there is an intricate web of distal basal notions that can be defined. But since most of them are not pertinent here, I defer discussing them to another occasion (Raven Reference Raven2021a). My focus here is on only the most pertinent distal basal notions.

The distal basal status of a fact is determined partly by the proximal basal statuses of other facts, such as its grounds. We assume that ground obeys a ‘cut’ principle that requires it to chain. We then define an auxiliary hereditary notion that includes all of a fact's grounds:

The hereditary basal status of an ungrounded fact collapses into its proximal basal status. The proximal and distal basal statuses of a fact cannot diverge if it is hereditarily (un)worldly.

But a less extreme distal notion allows for divergence by only requiring the existence of some hereditarily (un)worldly ground:

These distal notions follow proximal notions in regarding worldliness as dominant. Other distal notions could also be defined. My present concern, however, is not with these other notions and they are discussed elsewhere (Raven Reference Raven2021a).

We may illustrate the possibility of divergence with our earlier example, the fact ES. Presumably, the instantiation of the member of relation by Socrates and {Socrates} is unworldly. (If this is doubted, we can switch examples: the instantiation of the member of relation by 0 and {0} is unworldly.) If so, then Proximal-Atomic implies that ES is proximally unworldly. But it may also be supposed that set-theoretic facts are grounded in facts about mathematicians’ mental acts or practices. And it may be further supposed that these facts are proximally worldly. If so, then Distal implies that ES is distally worldly. This illustrates how a fact can be proximally unworldly but distally worldly.

The possibility of divergence between proximal and distal basal statuses helps resolve the puzzle in two stages. The first stage argues that the puzzle rests on a fallacy of equivocation or a false claim. Recall the claims generating the puzzle:

We did not distinguish between proximal/distal basal status before. Now that we do, does it affect these claims? It does not affect Social-Essentialism because it does not even mention basal statuses. But both Embedded and Detached do. They do not, however, specify whether the basal status is proximal or distal. They are ambiguous between them. One disambiguation has them be about proximal basal statuses:

Another disambiguation has them be about distal basal statuses:

The claims must be disambiguated uniformly to avoid the fallacy of equivocation. But the puzzle cannot arise either way. This is because either way will falsify some claim. However Detached is disambiguated, it will imply that essentialist facts are unworldly in the relevant sense. Because social-essentialist facts are essentialist facts, they, too, will be unworldly in that sense. And so, on any disambiguation, Embedded is false if Detached is true. Disambiguating uniformly thus guarantees that some claim will be false. So the puzzle cannot arise.

Although the first stage explains why the puzzle cannot arise, it does not explain why it appeared as if it could. The second stage explains this. The quick explanation is that any allure the puzzle appeared to have relied on equivocating over the proximal/distal distinction. This can be bolstered by a fuller explanation of why some of the disambiguated claims seem true even though they cannot together generate the puzzle. The explanation might take various forms. Because a comprehensive discussion of them is beyond the scope of this essay, below I focus on sketching one form of explanation that interprets detachedness proximally and embeddedness distally.

To interpret detachedness proximally is to regard essences' domain-fixing role as proximal. This role takes an essentialist fact to help fix what its subject is. Thus, the essentialist facts S1–S3 help fix what Congress is, what women are, and what the Korean DMZ is. None of this seems to turn on the worldly circumstances. Of course, one might still ask what, if anything, explains why these essentialist facts obtain. But how one answers does not seem relevant to their role in fixing what their subjects are. These considerations suggest that the detachedness of an essentialist fact is a proximal matter unconcerned with its grounds. That suggests that detachedness concerns proximal basal status. And that supports Detached-Proximal.

To interpret embeddedness distally is to regard the source of social essences to be distal. This relies on distinguishing between the role of essences and their ultimate sources. Even if it is then assumed that essences play a domain-fixing role, this won't determine what their ultimate source is. That, presumably, is determined distally by what, if anything, grounds the essentialist facts. But the grounds of essentialist facts are precisely what determines their distal basal status. And that supports Embedded-Distal.

In sum, my solution alleges that the puzzle equivocates the proximal/distal distinction. On the proper disambiguations, Detached says that essentialist facts are proximally unworldly whereas Embedded says that social-essentialist facts are distally worldly. Given Social-Essentialism, social-essentialist facts are proximally unworldly but distally worldly. These statuses are compatible. So contradiction is avoided.

This, of course, is only the beginnings of a complete solution. A proper defense of both stages is still needed. In particular, we will want fuller accounts of the proximal unworldliness of essentialist facts and of the distal worldliness of social-essentialist facts. And it may be challenging to provide such accounts (Raven Reference Raven2021b). But now at least there is a path laid out before us toward a solution to the puzzle.

The path seems to lead to a reconciliation of a conflict in our thought about social essences. The puzzle derived from our competing inclinations to regard social essences as at once detached from while embedded in the world. A proper defense of the solution just sketched, however, promises to vindicate the kernels of truth in each of these competing inclinations.

The path may also lead to a more general reconciliation of the Platonic and Aristotelian traditions. With the Platonist, essences may help fix their subject matter without regard to the circumstances. With the Aristotelian, essences, and social essences in particular, may nevertheless have their ultimate source in the worldly circumstances. It is understandable why these two traditions may at first seem to conflict. But my apparatus suggests how they might not conflict after all.